Estimating Structured Vector Autoregressive Models

Authors: Igor Melnyk, Arindam Banerjee

ICML 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on synthetic and real data with a variety of structures are presented, validating theoretical results.
Researcher Affiliation Academia Department of Computer Science and Engineering, University of Minnesota, Twin Cities
Pseudocode No The paper does not include a pseudocode block or clearly labeled algorithm.
Open Source Code No The paper does not provide any statement or link indicating the release of open-source code for the described methodology.
Open Datasets Yes We used the NASA flight dataset from (nas), consisting of over 100,000 flights, each having a record of about 250 parameters, sampled at 1 Hz. (nas) NASA Aviation Safety Dataset. Available at https://c3.nasa.gov/dashlink/projects/85/.
Dataset Splits Yes For each flight we separately fitted a first-order VAR model using five approaches and performed 5-fold cross validation to select λ, achieving smallest prediction error.
Hardware Specification No The paper does not specify the hardware used for running the experiments (e.g., GPU/CPU models, memory specifications).
Software Dependencies No The paper does not provide specific software dependencies with version numbers (e.g., specific libraries, frameworks, or solvers with their versions).
Experiment Setup Yes To evaluate the estimation problem with L1 norm, we simulated a first-order VAR process for different values of p [10, 600], s [4, 260], and N [10, 5000]. Regularization parameter was varied in the range λN (0, λmax)... For Sparse Group we set α = 0.5, while for OWL the weights c1, . . . , cp were set as a monotonically decreasing sequence.